Two identical transactions (same card, same amount, same customer) can cost you materially different amounts to process depending on the path they take: which acquirer accepts them, whether they're processed locally or cross-border, and which network carries them. Most merchants never see this spread, because their routing is static. Every payment goes down the same pipe agreed at onboarding, and the pricing conversation happens once a year, if at all.
Smart routing turns that one-time pricing decision into a per-transaction one. Before evaluating platforms, it's worth understanding exactly where the fee spread comes from, because that determines what a routing platform actually needs to be able to do.
Where processing fees actually vary
A card payment's cost has three main components: interchange (paid to the issuing bank), scheme fees (paid to the card network), and the acquirer's markup. Routing decisions influence all three.
- Acquirer choice per card type and transaction profile. Different acquirers price differently by card type, region, and volume tier. An acquirer that's cheap for domestic consumer debit may be expensive for commercial credit. With multiple acquirers connected, routing rules can send each transaction to whichever one prices that profile best, instead of accepting one blended rate across all traffic.
- Local acquiring vs. cross-border processing. When a transaction is acquired outside the cardholder's region, cross-border interchange uplifts and FX conversion costs apply. Routing European cards to a European acquirer entity, and processing in local currency, removes those uplifts entirely. This is one of the strongest structural levers available: it rewards merchants who can lean on local acquiring relationships market by market rather than forcing all traffic through one gateway.
Learn more: Local acquiring payments: What it is and how to set it up
- Debit network selection. In markets like the US, many debit cards can be routed over more than one network, and the eligible networks carry different interchange. A routing layer that selects the cheaper eligible network reduces cost with zero change to the customer experience.
- The hidden fee: declines. Every failed payment still incurs network and processing costs, and then costs you again in retries, support contacts, and lost sales. Recovering failures cheaply is part of the fee picture: revenue and fees already spent on a failed attempt are otherwise written off entirely.
What a smart routing platform needs to do
Judged against the mechanics above, a platform reduces fees only if it offers:
- Transaction-level rules, not static priority lists: conditions on card type, BIN country, currency, and amount, so each payment takes its cheapest eligible route.
- A broad, pre-built acquirer and processor network: cost-based routing is only as good as the options you can route between, and each new connection should be configuration, not an integration project.
- Cost visibility granular enough to act on: you can't route around fees you can't see. Effective cost per provider, per market, per card type.
- Authorization-rate protection: the cheapest route is worthless if it declines more. Fallbacks and per-route performance data keep cost optimization from cannibalizing conversion.
- Iteration without engineering: acquirer pricing and performance change; your rules need to change with them, ideally the same day.
Why merchants pick Primer to help cut processing fees
Primer's unified payment infrastructure addresses each requirement above directly:
- No-code rule building. Workflows is a no-code builder where payment teams define routing logic (by card type, geography, amount, or currency) and adjust it as pricing or performance shifts, without engineering tickets. Gaming marketplace Eldorado uses Workflows to A/B test payment configurations, an approach that applies equally to testing routing strategies against each other.
- Cost visibility built in. Costs Overview surfaces fee leakage and benchmarks providers against each other, giving the cost data that routing rules act on.
- Conversion protected alongside cost. Fallbacks ensure cost-optimized routing never comes at the expense of conversion: if a route fails, the transaction retries instantly on a backup processor. This is the mechanism behind crypto on-ramp Banxa recovering US$7 million in payments over 6 months, revenue that would otherwise have been written off along with the fees already paid to attempt it.
- Performance data across every provider. Observability standardizes performance data across every connected provider, so you can verify that a cheaper route is also converting, per market and per card type.
- Configuration, not integration. Because Primer sits above your processors rather than replacing them, adding an acquirer for a specific market or card type is a configuration exercise, not a new integration project.
See where your own fee spread is hiding. Book a demo and we'll walk through your routing setup with you.
FAQ
Does smart routing require changes to my checkout?
No. Routing logic operates behind the checkout, after the customer submits payment. Strategies can change without any customer-facing impact.
Can cost-based routing hurt authorization rates?
If configured on cost alone, yes, which is why per-route performance visibility and fallbacks matter. The goal is lowest cost among routes that convert, not lowest cost outright.
Is this only worthwhile at enterprise volume?
Absolute savings scale with volume, but any merchant with international traffic or more than one processor has a fee spread to capture. The structural levers (local acquiring, acquirer selection, decline recovery) apply at any size.
.png)
